模糊联想记忆神经网络模型在地震预报中的应用

介绍了模糊联想记忆FAM(Fuzzy Associative Memory)神经网络模型、FAM自适应学习算法以及FAM推理机的原理，并成功地将其用于新一代的地震预报专家系统NGESEP，使得系统既具有良好的学习功能，又避免了通常神经网络学习知识隐含在权值中不易被人们理解和专家系统解释的缺点。      

地震预报专家系统ESEP 3.0

近期研制成功的地震预报专家系统ESEP 3．0将模糊系统、神经网络与专家系统技术相结合，引入了驾驭式的推理机制，除具有第一代专家系统的符号推理与解释功能、以及第二代专家系统的学习功能外，还具有较强的人机交互能力，是一个全新的专家系统。ESEP 3．0由知识编辑、机器学习、驾驭式模糊推理机和解释等4个子系统组成，本文介绍了系统的组成和概况。     

震后趋势决策支持系统PTDSS

介绍了作者研制的“震后趋势决策支持系统PTDSS”的构成及预报效果,该系统使用模糊联想记忆FAM（Fuzzy Associative Menory)模型将神经网络与模糊系统结合起来,使之具有第2代专家系统的学习功能,系统知识库中除存有专家多年来总结的有关地震现场预报的知识外,还通过系统学习得到了一批十分有意义的,目前专家尚未总结的知识,震例检验的结果该系统有较好的预报效果。     

基于减法聚类模糊神经网络的砂土液化势判别

砂土地震液化问题是岩土地震工程学的重要研究课题之一。在分析模糊神经网络原理的基础上,利用减法聚类算法对自适应模糊推理系统进行优化,并建立了砂土地震液化的模糊神经网络模型。然后,将该模型用于实际工程的砂土液化判别中,并与传统砂土液化判别方法结果进行对比。判别结果表明:文中建立的模糊神经网络模型具有较强的学习功能,用于砂土地震液化判别中是可行的和有效的。     

洪水预报中特征值预报的若干数学方法比较

讨论研究了水文特征值预报的数学方法,统计回归模型、神经网络模型和模糊回归模型。三个计处实例表明如果系统的线性关系较好,统计回归模型的结果最好;如果系统的线民生关系差,神经网络模型的结果最好;如果用于率定模型的资料太短,任何一个模型都不可靠。     

径向基函数(RBF)神经网络及其应用

介绍了径向基函数(RBF)神经网络的原理、 学习算法及其在地震预报专家系统ESEP 3.0中的应用。 实际应用结果表明, 该神经网络可以很好地克服BP神经网络学习过程的收敛过分依赖于初值和可能出现局部收敛的缺陷, 具有较快的运算速度、 较强的非线性映射能力和较好的预报效能。     

BP神经网络在新一代地震预报专家系统中的应用

简介了新一代地震预报专家系统ＮＧＥＳＥＰ，ＢＰ神经网络模型及其算法，同盱ＢＰ神经网络具有很强的非线怀映射功能，它可以很好地反映震前出现异常的种类和异常时间与未来地震震级之间的较强非线性关系，在ＮＧＥＳＥＰ系统中可以从实例库中提取典型震例并通过ＢＰ网络进行学习，实际震例检验表明系统对未来地震震级的预测取得较好理想的结果。     

烈度评定专家系统——EIE

本文详细地介绍了地震烈度评定专家系统—EIE。地震烈度的评定是一件非常重要而又十分经常的工怍。由于它包含了许多模糊概念,而评定的合适与否又在很大程度上依赖于人们的经验,因此运用人工智能的理论与方法,作成相应的专家系统是非常必要的。地震烈度评定专家系统—EIE的主要特点是运用了落影理论和模糊推理,从而很好地处理了糊模信息。本文对专家系统EIE的框架、知识库,推理机均作了详细介绍。本文的目的既是向地震和地震工程界呈献一个有用的专家系统,又是对人工智能在地震工程中的应用起一个推动作用。特别是对落影理论和模糊推理在专家系统中的应用,本文提出了一个有普遍意义的框架。     

地震预报专家系统ESEP 3.0中的知识表示方法

知识表示是专家系统建造中的一个关键环节，知识表示方法的好坏直接关系到系统的智能水平和解决问题的能力。在地震预报专家系统ESEP3．0中有多种知识表示方法，本文介绍了该系统的结构知识、地震学和前兆预报知识、机器学习知识、综合预报知识、确定异常证据可信度知识以及支持知识等多种知识表示方法，提出了确定异常证据可信度模型。这些知识表示方法可以很好地表示各类地震预报知识。     

新一代的地震预报专家系统NGESEP中的推理子系统

本文介绍了第一代专家系统和第二代专家系统的特点以及新一代地震预报专家系统中的推理子系统。该系统根据地震预报知识的特点设计了推理机,解决了推理子系统中的一系列问题。     

The FAM(Fuzzy Asociative Memory)neural network model and its application in earthquake prediction

ＴｈｅＦＡＭ（ＦｕｚｚｙＡｓｏｃｉａｔｉｖｅＭｅｍｏｒｙ）ｎｅｕｒａｌｎｅｔｗｏｒｋｍｏｄｅｌａｎｄｉｔｓａｐｐｌｉｃａｔｉｏｎｉｎｅａｒｔｈｑｕａｋｅｐｒｅｄｉｃｔｉｏｎＷＥＩＷＡＮＧ１）（王炜）ＧＥＮＧＦＥＮＧＷＵ２）（吴耿锋）ＢＩＮＳＨＵＨ...     

新一代地震预报专家系统NGESEP预报效能的初步研究

利用４０个地震震例对新一代地震预报专家系统ＮＧＥＳＥＰ的预报效果进行了检验。比较了使用信度分布图和信度等值线图两种方法未来发震地点的效能，分析了系统对地震三要素的预报结果，表明该系统具有较高的预报效能。     

基于动态模糊神经网络的结构主动控制仿真分析

在模糊控制中,如何更加合理地生成控制规则,是其应用的一个重要问题。本文采用动态模糊神经网络(DFNN)算法,并借助于最优控制算法的样本数据,实现建筑结构振动控制中的模糊规则自动提取。首先,介绍了DFNN的结构和算法;其次,采用DFNN算法设计了二输入单输出及四输入单输出两种模糊控制器,对顶层设置AMD控制装置的五层钢框架模型结构进行模糊控制仿真分析。仿真结果表明,两种模糊控制器对顶层位移和加速度反应峰值的控制效果达到50%和30%以上,对地震输入和结构参数的变化均具有较好的鲁棒性;相比二输入模糊控制器,四输入模糊控制器的控制效果更好。本文研究为地震作用下建筑结构AMD模糊控制提供了新的思路和方法。     

Fuzzy neural network models for liquefaction prediction

Integrated fuzzy neural network models are developed for the assessment of liquefaction potential of a site. The models are trained with large databases of liquefaction case histories. A two-stage training algorithm is used to develop a fuzzy neural network model. In the preliminary training stage, the training case histories are used to determine initial network parameters. In the final training stage, the training case histories are processed one by one to develop membership functions for the network parameters. During the testing phase, input variables are described in linguistic terms such as ‘high’ and ‘low’. The prediction is made in terms of a liquefaction index representing the degree of liquefaction described in fuzzy terms such as ‘highly likely’, ‘likely’, or ‘unlikely’. The results from the model are compared with actual field observations and misclassified cases are identified. The models are found to have good predictive ability and are expected to be very useful for a preliminary evaluation of liquefaction potential of a site for which the input parameters are not well defined.     

A simulation-based fuzzy chance-constrained programming model for optimal groundwater remediation under uncertainty

In this study a simulation-based fuzzy chance-constrained programming (SFCCP) model is developed based on possibility theory. The model is solved through an indirect search approach which integrates fuzzy simulation, artificial neural network and simulated annealing techniques. This approach has the advantages of: (1) handling simulation and optimization problems under uncertainty associated with fuzzy parameters, (2) providing additional information (i.e. possibility of constraint satisfaction) indicating that how likely one can believe the decision results, (3) alleviating computational burdens in the optimization process, and (4) reducing the chances of being trapped in local optima. The model is applied to a petroleum-contaminated aquifer located in western Canada for supporting the optimal design of groundwater remediation systems. The model solutions provide optimal groundwater pumping rates for the 3, 5 and 10 years of pumping schemes. It is observed that the uncertainty significantly affects the remediation strategies. To mitigate such impacts, additional cost is required either for increased pumping rate or for reinforced site characterization.     

Conceptual fuzzy neural network model for water quality simulation

Artificial neural networks (ANNs) have been applied successfully in various fields. However, ANN models depend on large sets of historical data, and are of limited use when only vague and uncertain information is available, which leads to difficulties in defining the model architecture and a low reliability of results. A conceptual fuzzy neural network (CFNN) is proposed and applied in a water quality model to simulate the Barra Bonita reservoir system, located in the southeast region of Brazil. The CFNN model consists of a rationally‐defined architecture based on accumulated expert knowledge about variables and processes included in the model. A genetic algorithm is used as the training method for finding the parameters of fuzzy inference and the connection weights. The proposed model may handle the uncertainties related to the system itself, model parameterization, complexity of concepts involved and scarcity and inaccuracy of data. The CFNN showed greater robustness and reliability when dealing with systems for which data are considered to be vague, uncertain or incomplete. The CFNN model structure is easier to understand and to define than other ANN‐based models. Moreover, it can help to understand the basic behaviour of the system as a whole, being a successful example of cooperation between human and machine. Copyright © 2006 John Wiley & Sons, Ltd.